Int. J. Plant Sci. 165(3):437–444. 2004.
Ó 2004 by The University of Chicago. All rights reserved.
1058-5893/2004/16503-0010$15.00
GENOTYPIC AND GENETIC DIVERSITY OF THE COMMON WEED
CIRSIUM ARVENSE (ASTERACEAE)
Magali Solé,1 ,* Walter Durka,* Sabine Eber,y and Roland Brandlz
*Department of Community Ecology, Centre for Environmental Research Leipzig-Halle, Theodor-Lieser Strasse 4, 06120 Halle, Germany;
yNatural Environment Research Council Centre for Population Biology, Imperial College at Silwood Park, Ascot,
Berkshire SL5 7PY, Great Britain; and zUniversity of Marburg, Department of Animal Ecology,
Karl-von-Frisch Strasse, D-35032 Marburg, Germany
In many clonal species, seedling establishment is restricted to early successional stages when recruitment is
still possible. Then, one expects that adapted genotypes become dominant and genotypic and genetic diversity
should decrease with time. We investigated genotypic and genetic diversity within recently founded and
established populations of the common weed Cirsium arvense. We used highly polymorphic amplified
fragments length polymorphism (AFLP) markers. All populations were multiclonal and highly diverse (the
proportion of distinguishable genotypes was 0:73 6 0:25 [mean 6 SD]). Clonal evenness was variable and
ranged from 0.2 to 1. Independent of successional stage, we found on the small geographic scale of our study
(<5 km) a considerable differentiation between populations (FSC ¼ 0:63). This amount of differentiation was
similar between founder and established populations and could result from selection in the early stage of
succession as well as founder effects. Contrary to the general expectation, genotypic and genetic diversity were
maintained through time, and molecular variance did not differ between successional stages (1:9 6 0:89 vs.
2:5 6 1:41). We suggest that this pattern is a consequence of the particular reproductive system of C. arvense
that combines clonality with dioecy. The combination of clonal reproduction with the recruitment of sexually
outcrossed seedlings in the first years allows the species to perform efficient colonizations even with founder
effects, to undergo selection without loss of diversity, and to persist locally. This strategy appears to be very
efficient in C. arvense and may have contributed to the worldwide success of this species.
Keywords: amplified fragments length polymorphism (AFLP), Cirsium arvense, clonal plant, genotypic
diversity, molecular variance, population differentiation, succession.
Introduction
with gene flow and recombination will act both on the genetic
variability within and among populations.
Cirsium arvense is a long-lived perennial weed that is
abundant on agricultural land. It has a mixed sexual-asexual
reproduction system (Lalonde and Roitberg 1994). It reproduces vegetatively by means of very efficient lateral roots that
ensure a rapid expansion of ramets. Moore (1975) reported
two independent studies that found a spread by lateral roots
of 6 m per season, whereas Bostock and Benton (1979) in
another study noted a spread of even 12 m per year. Cirsium
arvense is often described as a subdioecious-dioecious species. Female plants are strictly female. Male plants possess
vestigial ovaries and are morphological hermaphrodites but
are considered to be functional males (Moore 1975). Cirsium
arvense is insect pollinated and produces plumed achenes
with a long pappus. Natural long-distance dispersal by vegetative propagules is limited (unless roots are transported by
humans, e.g., in soil); hence, new populations must be established using seeds.
In plant species with a dioecious breeding system, the genotypic diversity of seedlings establishing a new population
is expected to be high compared with other less outcrossed
species. However, after the establishment of seedlings, genotypes with vigorous clonal growth may be favored, as clonality is an efficient strategy to spread within a habitat. Hence,
Clonal plants can reproduce by sexual and asexual reproduction. Whereas sexual reproduction accounts for recombination and dispersal, clonality propagates the same genotype
locally. Therefore, it is often suggested that the spatial distribution of genotypic and genetic diversity within and across
populations reflects the balance between clonal growth and
successful sexual reproduction (Sackville Hamilton et al.
1987; Schmid 1990; McLellan et al. 1997). Alongside these
two modes of reproduction, it is important to distinguish between genotypic and genetic diversity. Genotypic diversity is
usually called clonal diversity in clonal plants; it stands for
the number of genotypes within populations, whereas genetic
diversity represents the variability between genotypes. Genetic
and genotypic diversity is influenced by a number of processes
(e.g., mode of reproduction, gene flow, mutation, selection).
Clonal reproduction together with selection and mutation will
principally affect the genotypic diversity and spatial distribution of clones within populations, whereas sexuality together
1
Author for correspondence; e-mail [email protected].
Manuscript received February 2003; revised manuscript received December
2003.
437
438
INTERNATIONAL JOURNAL OF PLANT SCIENCES
after some time only a few genotypes should dominate the
habitat due to competitive exclusion (Gray 1987; Eriksson
1993). This process should then lead to a decrease in genotypic diversity when populations are aging, i.e., in later successional stages.
The temporal development of genotypic diversity has already been studied in other herbaceous perennials. Studies
that followed genotypic diversity through time (Hartnett and
Bazzaz 1985) or compared genotypic diversity between sites
that differed in successional stage (McNeilly and Roose 1984;
Maddox et al. 1989) showed a decline in the number of
genets over time. Studies that compared populations of clonal
plants in habitats with different disturbance regimes reached
similar conclusions. Populations in unstable habitats tended
to have a higher genotypic diversity compared with populations in stable habitats (Piquot et al. 1996; Xie et al. 2001).
However, in some cases no decrease in genotypic diversity was
found (Verburg et al. 2000). Some studies that followed genetic diversity through time with quantitative markers supported the theoretical predictions of a decrease in genetic
diversity with time (Aarssen and Turkington 1985). In others,
however, no decrease in genetic diversity over time was observed (Hartnett et al. 1987; Taylor and Aarssen 1988).
Our study is based on molecular data. We analyzed genotypic and genetic diversity within recently founded as well as
established populations of C. arvense. Because of its high
power of resolution in distinguishing genotypes, we used amplified fragment length polymorphism (AFLP) (Vos et al.
1995). Our article aims to address the following questions:
(1) Is there a decrease in genotypic diversity and evenness of
C. arvense populations with time? and (2) How is the genetic
variation partitioned within and among populations?
Material and Methods
All sampled populations are located within a 15-km2 rural
area in southern Germany (11°509E, 49°359N). This area is a
complex mosaic of roadsides, meadows, cultivated and abandoned agricultural fields, meadows, roadsides, and wasteland.
In September 1999, samples were collected from 16 populations of Cirsium arvense from two contrasting successional
stages: founder and established populations. We identified
each successional stage according to information available
from previous studies (Eber and Brandl 2003) and ecological
criteria. Founder populations of C. arvense occurred in habitats where the vegetation was in a typical early successional
stage. The vegetation cover was <75% and the plant community was exclusively composed of herbaceous species.
Cirsium arvense was the dominant plant species. Shoots of
C. arvense were easily recognizable as freshly germinated.
Moreover, during our previous attempts to map all populations of C. arvense in the same area, these populations did
not exist (Eber and Brandl 2003). Thus, we are sure that
these populations date from the year of our study. Because of
the difficulty in finding founder populations, we were able
to sample only six populations. Established populations of
C. arvense were sampled in old fallows. Within theses communities, C. arvense populations were in an advanced or regressive successional phase. There, the vegetation cover was
totally closed and woody species were present. Cirsium
arvense shoots were particularly high and ligneous. All these
established populations had already been mapped at this
advanced stage in 1994. Hence, populations were an age of
at least 6 yr old.
In this article, we define the population size as the number
of shoots within populations. In 1999, the population size of
all 16 populations was estimated. For populations with <100
shoots, the number of shoots was counted and rounded off
to the nearest ten. For bigger populations, the average number of plants was counted within randomly placed 1-m quadrates. Counts were extrapolated to the population area
covered by C. arvense, with the number of shoots being
rounded off to the nearest hundred or thousand.
Populations differed markedly in size, as well as in density
and patchiness of shoots. Therefore, we could not apply
a regular spatial sampling design. Moreover, because we
were interested in estimating the genotypic diversity of C.
arvense rather than the spatial extent of certain genotypes,
we adopted a random sampling strategy. We covered each
population and randomly sampled fresh leaves of C. arvense
shoots. In order to avoid sampling the same genet twice, we
limited our sampling within a patch; i.e., for an equal number of shoots, patchy populations were less sampled than
uniform populations (e.g., E10 and E2). Thus, sampling effort reflects the population area and patchiness. Because of
their small size (n ¼ 10) we sampled almost all shoots in
populations E7 and F4 (eight and seven, respectively) in order to obtain a fairly accurate value of the genotypic diversity in these populations.
The number of plants analyzed was not correlated with
the population size. In order to test for a potential sampling
artifact on our results, we repeated the molecular statistical
analyses three times by taking only eight random samples
from each population (except for populations F1 and F4,
where seven samples were available). These analyses with
a restricted sample size led to similar results as those from
the analysis with the entire sample size (data not shown).
We extracted DNA from young leaves with CTAB (Doyle
and Doyle 1988). DNA quality and concentration were estimated from 5.5 mL of the extract on an 0.8% agarose gel.
For AFLP (Vos et al. 1995) we used 0.5 mg of DNA per sample. We followed the Ligation and Preselective Amplification
Modules for Small Plant Genomes procedure from Applied
Biosystems except that the digestion and the ligation were
performed in a MWG-Biotech Primus 96 thermocycler at
37°C for 2 h. An initial screening using 64 selective primer
combinations was performed on a random sample of 10 individuals across all sampled populations. From that analysis
the two primer combinations EcoRI-ACC/MseI-CTG and
EcoRI-ACG/MseI-CTT appeared to be sufficiently polymorphic to discriminate clones within populations. Fragments
were separated on an ABI PRISM 310 genetic analyzer with
100 units as a minimum height threshold for peak detection.
Data were imported to the Genotyper analysis software, but
we were not able to genotype our data automatically because
of the strong sensitivity of the Genotyper program and the
heterogeneity among runs. Such lane-to-lane variation has already been observed in previous studies (De Riek et al.
1999). Following De Riek et al., we only used the Genotyper
SOLÉ ET AL.—GENOTYPIC AND GENETIC DIVERSITY IN CIRSIUM ARVENSE
program to produce a preliminary presence/absence matrix,
which was subsequently checked manually. For the present
study we used 93 polymorphic loci, 42 for EcoRI-ACC/MseICTG and 51 for EcoRI-ACG/MseI-CTT. Samples showing
the same multilocus AFLP phenotype were considered to be
the same genotype.
The AFLP data were analyzed at two hierarchical levels:
within and among populations. As an estimator of the intrapopulation genotypic diversity we used the proportion of distinguishable genotypes (i ¼ G=n; Ellstrand and Roose 1987),
where G is the number of genotypes and n the number of
sampled shoots. Clonal evenness was calculated as the relative abundance of each genotype within a population. A
number of evenness indices are available, and there is no consensus on which one is the best (Smith and Wilson 1996).
We chose the evenness index E1=D ¼ 1=D
G (Williams 1964),
2
which is based on Simpson’s index (D ¼ G
i¼1 pi , where pi is
the relative abundance of the ith genotype); E1/D ranges from
0 to 1. High values characterize populations with an even
distribution of clones; low values characterize populations
dominated by only a few genotypes. We compared the genotypic diversity and the evenness index between the two successional stages with a Mann-Whitney U-test.
For each population we calculated the molecular variance
as the average number of mismatches of bands within a population (sum of mismatches divided by 2N (N 1), where N
equals the number of samples), and we tested for correlations
between the molecular variance and population size. We
tested for correlations between the population size and the
three factors molecular variance, genotypic diversity, and
clonal evenness, using Spearman’s rank correlations.
To study the partitioning of genetic variance among populations, we conducted an analysis of molecular variance
(AMOVA; Excoffier et al. 1992) with the program Arlequin
(Schneider et al. 2000). We constructed a hierarchical model
with genotypes nested within populations and populations
nested within successional stages (table 1). Variance components (s2a , s2b , s2c ; see table 2 for explanations), the sum of all
439
squared differences, and analogues of F-statistics (F) were calculated. The significance of estimated parameters was tested
by a permutation procedure (Excoffier et al. 1992). The parameters FCT and s2a were tested by random permutation of
genotypes of whole populations across successional stages. In
this study FCT estimated the successional stage effect. The parameters FSC and s2b were tested by random permutation of
individuals across populations but within the same successional stage. In this study FSC estimated the population differentiation and is the equivalent of the Wright’s FST index
(Wright 1965). Finally, FST and s2c were tested by randomly
permuting AFLP phenotypes among populations and between
successional stages. We used a matrix correlation test with
1000 permutations to test whether the matrix of genetic distances (FSC =ð1 FSC ); Rousset 1997) was correlated with the
matrix of geographic distances (table 3). The genetic distance
matrix was calculated by the program Arlequin as a pairwise
population FSC matrix (Schneider et al. 2000).
Results
From the AFLPs we identified 93 polymorphic loci. Across
the 307 sampled Cirsium arvense shoots, we distinguished
231 haplotypes (clones); 86% of these clones were found only
once. Genotypes shared by several individuals (14%) always
occurred in the same population. The mean genotypic diversity over all populations was i ¼ 0:73 6 0:25 (mean 6 SD).
However, genotypic diversity differed considerably between
populations (range: 0.25–1.0; table 1). The clonal evenness
(E1=D ) was also very variable and ranged from 0.2 to 1 (mean
E1=D ¼ 0:6 6 0:32). Both indices (i and E) were highly correlated (Spearman’s rank correlation coefficient rs ¼ 0:98,
P < 0:0001). Genotypic diversity and the clonal evenness did
not differ significantly between founder and established populations (genotypic diversity: U ¼ 24, P > 0:3; evenness:
U ¼ 26:5, P > 0:3) and were not correlated with the population size (genotypic diversity: r ¼ 0:1; P > 0:70; evenness
r ¼ 0:03; P > 0:92).
Table 1
Demographic, Genotypic, and Genetic Characteristics of 16 German Populations of the Perennial Dioecious Cirsium arvense
Population
name
Successional
stage
F1
F2
F3
F4
F5
F6
E1
E2
E3
E4
E5
E6
E7
E8
E9
E10
Founder
Founder
Founder
Founder
Founder
Founder
Established
Established
Established
Established
Established
Established
Established
Established
Established
Established
Population
size
Area (m2)
No. of plants
analyzed (n)
No. of genotypes
detected (G)
Genotypic
diversity (i)
Evenness
index (E1/D)
Molecular
variance
30
40
200
10
100
120
500
100
30
3000
400
40
10
800
8000
100
6
25
40
10
80
50
750
160
200
1200
300
50
10
130
1000
80
7
15
17
7
19
10
53
20
20
54
28
12
8
11
18
8
6
7
16
7
9
10
42
12
20
52
13
6
2
11
13
5
0.86
0.47
0.94
1.00
0.47
1.00
0.79
0.60
1.00
0.96
0.46
0.50
0.25
1.00
0.72
0.63
0.78
0.28
0.89
1
0.2
1
0.58
0.29
1
0.93
0.29
0.33
0.2
1
0.5
0.5
2.12
0.48
4.59
3.45
1.78
2.63
2.48
1.93
3.01
2.92
1.07
1.06
0.75
2.42
0.76
2.39
INTERNATIONAL JOURNAL OF PLANT SCIENCES
440
Table 2
Hierarchical Analysis of Molecular Variance Testing for Differentiation between Successional Stages,
among Populations within Successional Stages, and within Populations
Variation component
Between successional stages
Among populations within
successional stages
Within populations
Variance
% Total
Significance
F-Statistics
s2a
0.58
5
P < 0.01
FCT = 0.05
s2b
s2c
7.46
4.44
60
35
P < 0.001
P < 0.001
FSC = 0.63
FST = 0.64
Note. Significance tests are based on 1000 permutations.
with other data from the literature are difficult to interpret
because reviews are based on different sample sizes, spatial
scales, and molecular markers and compare plant species
with different life and phylogenetic histories (Gitzendanner
and Soltis 2000). However, compared to the most widely
cited reviews of genetic diversity patterns in clonal plants
(Ellstrand and Roose 1987; Widen et al. 1994), average genotypic diversity in C. arvense is much higher (0.73 vs.
0.17). This big difference found with AFLP and microsatellites might result from the higher power of resolution of
the DNA-based molecular markers, whereas the abovementioned reviews are mostly based on allozyme studies,
which offer fewer loci. Nevertheless, the high genotypic diversity found in C. arvense is confirmed by the fact that
100% of the sampled populations were multiclonal, compared to an average of 62% in Ellstrand and Roose (1987)
and Widen et al. (1994).
Like genotypic diversity, the clonal evenness may vary
greatly among clonal species. Populations can be monoor multiclonal (Aspinwall and Christian 1992; Eckert and
Barrett 1993; McClintock and Waterway 1993; Piquot et al.
1996). For multiclonal populations, almost all patterns of
genotypic diversity combined with clonal evenness can be
found (Ayres and Ryan 1997; Gabrielsen and Brochmann
The average molecular variance was 2.1 (61.1). Established populations had a lower mean molecular variance than
the founder populations (1:9 6 0:89 vs. 2:5 6 1:41), but the
difference was not significant (U ¼ 23, P > 0:44). Molecular
variance was not correlated with the population size (fig. 1).
Most of the haplotype diversity was found among populations within successional stages (s2b ¼ 60%), which results in
a high FSC value (0.63). Nevertheless, a considerable amount
of diversity was found within populations (s2c ¼ 35%). The
variance from differentiation between founder and established populations was significant but explained only a small
part of the total variance (s2a ¼ 5%).
Pairwise genotypic distances were not correlated to geographic distances (matrix correlation ¼ 0:14, P > 0:13; fig. 2).
Discussion
Genotypic diversity and clonal evenness of Cirsium arvense
were high and varied greatly among the 16 sampled populations. These results agree with recent data based on microsatellites. Jump et al. (2003) found an average genotypic
diversity of 0.68 (ranging from 0.08 to 0.97) and an average
evenness of 0.68 (ranging from 0 to 1). Direct comparisons
Table 3
Matrices of Pairwise Genetic and Geographic Distances among 16 German Populations of Cirsium arvense
F1
F1
F2
F3
F4
F5
F6
E1
E2
E3
E4
E5
E6
E7
E8
E9
E10
0.87
0.34
0.53
0.71
0.61
3.13
3.15
1.25
1.58
2.80
2.65
2.60
0.63
2.28
2.23
F2
F3
F4
F5
F6
E1
E2
E3
E4
E5
E6
E7
E8
E9
E10
0.13
1.83
1.80
0.73
0.75
2.28
0.85
0.75
2.48
1.25
0.88
0.83
2.03
1.53
0.63
0.70
0.62
0.53
0.56
0.63
0.55
0.72
0.76
0.46
0.52
0.56
0.66
0.56
0.60
0.70
0.44
0.50
0.64
0.57
0.62
2.15
0.59
0.58
0.43
0.47
0.60
0.50
0.49
1.80
0.38
0.82
0.85
0.58
0.74
0.78
0.77
0.64
0.66
0.73
0.63
0.80
0.87
0.53
0.66
0.77
0.74
0.60
0.72
0.65
0.60
0.28
0.83
0.94
0.65
0.79
0.85
0.81
0.77
0.82
0.75
0.72
2.18
2.13
2.95
3.05
1.68
3.00
3.75
3.48
4.13
0.65
2.15
1.88
2.83
2.75
0.63
0.80
0.88
0.57
0.73
0.77
0.77
0.65
0.75
0.67
0.64
2.20
1.75
3.43
0.72
0.69
0.82
0.43
0.43
0.68
0.62
0.60
0.60
0.49
0.54
1.70
1.50
3.20
0.55
0.25
0.55
0.77
0.80
0.73
3.15
3.25
1.35
1.68
2.98
2.73
2.73
0.77
2.25
2.25
0.22
0.44
0.35
2.58
1.55
0.63
0.28
1.55
1.35
1.10
0.45
2.35
2.13
0.43
0.51
3.85
3.35
1.78
2.05
3.55
3.33
2.83
0.53
3.00
2.95
0.64
3.08
3.90
1.85
2.20
3.25
3.00
3.38
0.71
2.08
2.13
2.48
3.50
1.40
1.75
2.65
2.38
3.00
0.60
1.50
1.53
3.80
2.48
2.50
1.50
1.45
3.53
0.61
1.05
0.95
0.58
0.51
2.35
2.35
0.53
0.62
3.80
3.55
0.50
1.80
1.53
1.63
0.57
1.98
1.80
1.60
1.38
1.65
0.56
2.15
1.95
0.78
0.90
0.70
0.82
0.79
0.89
0.71
0.83
0.71
0.82
0.91
0.79
4.00
3.75
0.79
Note. Genetic distances (mean number of pairwise mismatches) are presented in the lower part of the table; geographic distances (km) are
in the upper part.
SOLÉ ET AL.—GENOTYPIC AND GENETIC DIVERSITY IN CIRSIUM ARVENSE
Fig. 1 Population size (number of shoots counted per population)
in 16 German populations of Cirsium arvense was not correlated with
the molecular variance (rs ¼ 0:09, P > 0:72).
1998; Ivey and Richards 2001b). In C. arvense, the genotypic
diversity is high, and populations are not dominated by one
clone, a pattern that also was found in other species (Chung
and Epperson 1999; Auge et al. 2001; Xie et al. 2001).
Population differentiation frequently has been analyzed
with allozymes. Three main results emerged: differentiation
between populations is rather similar in plant species with
mixed and purely sexual reproduction (GST values 0.21 and
0.23, respectively; Hamrick and Godt 1990); most variation
occurs within populations (Baur and Schmid 1996); and population differentiation is lower in outbreeding than in inbreeding species (GST values <20% and >50%, respectively).
Studies based on dominant markers (RAPD) lead to the same
conclusions (Bussel 1999). In C. arvense most of the genetic
variation was among populations (60% of the total variation). All haplotypes were strictly local. The FSC -statistic
(equivalent in our study to Wrights’s FST value) was very high
(FSC ¼ 0:64), even compared with previous studies (0.25 in
Jump et al. 2003).
At a landscape scale, we found that differentiation among
populations was not correlated to geographic distance. These
results fortify those of Jump et al. (2003), who observed an
isolation by distance in C. arvense occurring from 200 km. A
high differentiation among populations independent of geographic distance, like in C. arvense, is more common in rare,
partially selfing, locally dispersed, or gene flow limited species (Travis et al. 1996; Fischer and Matthies 1998; Schmidt
and Jensen 2000). Thus, the high population differentiation
independent of geographic distance in C. arvense is surprising because C. arvense is an abundant outbreeding species.
In plants with mixed reproduction systems, clonal propagation can bias the estimation of F-statistics (McLellan et al.
1997). For example, in the clonal Cladium jamaicense the
441
FST based on the ramet (i.e., all aerial shoots coming from
the clonal propagation of a single root) was 0.68, whereas
the FST based on the genets (genetic individuals) was 0.035
(Ivey and Richards 2001a). In C. arvense, the analysis on the
genet level also resulted in a high population differentiation
(FSC ¼ 0:53, P < 0:001). Therefore, clonality is not the reason for population differentiation, a result already evident
from the fact that populations were not dominated by one or
a few clones. In addition to clonality, high population differentiation independent of geographic distance can also result
from strong selection (Endler 1986) as well as from founder
effects acting together with drift, low gene flow among populations, or low seedling recruitment within populations
(Slatkin 1977; Whitlock 1992).
Similarity among genotypes within populations, and thus
dissimilarity among populations, is expected to increase in
small and isolated populations as a consequence of genetic
drift, bottlenecks, and inbreeding (Hartl and Clark 1989;
Barrett and Kohn 1991). A decline in molecular variance in
small populations was frequently found, e.g., in the nonclonal Gentianella germanica (Fischer and Matthies 1997) as
well as in the clonal herb Ranunculus reptans (Fischer et al.
2000). In our study, population size was correlated neither
with molecular variance nor with genotypic diversity. The
correlation performed on established populations only, which
are more likely to have experienced genetic drift, was also
not significant. Thus, genetic drift does not seem to play an
important role in population differentiation.
Gene flow among populations can be distinguished into
pollen and seeds. The pollinators of C. arvense, mainly bumble bees, are very mobile (Walther-Hellwig and Frankl 2000).
The observation that seeds were produced in all populations
Fig. 2 Genetic and geographic distances in 16 German populations
of Cirsium arvense were not correlated (matrix correlation ¼ 0:12,
P > 0:18). High population differentiation observed in C. arvense is
not due to geographic distance.
442
INTERNATIONAL JOURNAL OF PLANT SCIENCES
even with 100% of females (M. Solé, unpublished manuscript) and the fact that apomixis does not occur (Correns
1916) confirm an efficient transport of pollen among populations. However, low seed dispersal and absence of seedling
recruitment in C. arvense are potential causes that generate
and maintain differentiation. Despite producing plumed
seeds, seed dispersal in C. arvense may not be very effective.
The pappus often breaks off the seed, so that at a distance of
1000 m only 0.2% of trapped pappi carried a seed (Bakker
1960; Bostock and Benton 1979). Like in other clonal plants
(Eriksson 1992; Wolf et al. 2000), the recruitment of C.
arvense seedlings is possible only during the early phases of
colonization. Cirsium arvense seedlings are very susceptible
to shading and competition (Bakker 1960). Our observations as well as reports in the literature indicate that the
recruitment of C. arvense seedlings is not possible in natural
or artificial plant communities with a dense plant cover
(Bostock and Benton 1983; M. Solé, unpublished data).
Thus, once populations have reached a closed canopy, seedling recruitment will stop. This phenomenon could then reinforce the population differentiation.
Two potential driving forces remain to explain the surprisingly high genetic differentiation among C. arvense populations: founder effects and selection. We tested for founder
effects (i.e., restricted-source origin of founder populations) by
comparing the level of genetic differentiation among founder
and established populations (Whitlock and McCauley 1990).
The FSC values were 0.55 (P < 0:001) for founder populations versus 0.64 (P < 0:001) for established populations. As
differentiation among founder populations was already high,
this result may indicate the foundation of new populations by
a nonrandom sample of the propagule pool of C. arvense.
In C. arvense, founder effects and selection are not mutually exclusive. However, founder effects and strong selection
usually tend to decrease the within-population genetic variability (Endler 1986; Pannell and Charlesworth 1999). In our
study, genotypic and genetic diversity did not decrease from
founder to established populations. We offer the following
explanation. We sampled plants in September, and thus the
founder populations already passed the seedling stage. If
mortality and strong selection occur during the seedling
stage, our sampling scheme was not able to retrieve the decrease of genetic diversity because the founder populations
had already passed the filter of a first selection phase. As long
as the selection regime is not spatially autocorrelated, this
will lead to high differentiation among populations.
However, the maintenance of genetic diversity through
time could also be fostered by the obligatory outcrossed
breeding system of C. arvense. Comparable results (high genetic diversity coupled with a high differentiation of populations) were already found in other clonal dioecious species
(Sherman-Broyles et al. 1992 in Rhus species; Gibson and
Wheelwright 1995). Because of dioecy, C. arvense seeds are
strictly outcrossed and thus must be highly variable. Their
recruitment during the early stages of succession could
compensate for the loss of diversity within a population.
Therefore, seedling recruitment in C. arvense might be an
important mechanism for initiating and maintaining a high
level of neutral genetic variability, as already suggested by
Heimann and Cussans (1996). Recently, theoretical studies
about the genetic diversity in clonal plants led to similar conclusions (Bengtsson 2003). In his model, Bengtsson looked at
the ‘‘genotypic identity’’ of populations (i.e., the probability
that two randomly sampled adult individuals from a population have the same genotype) depending on the population
growth parameters, rate of sexuality, and recruitment of sexually derived offspring. From his simulations, Bengtsson proposed that clonal populations possess an effective ‘‘memory’’
of their earlier genetic history, in the way that ‘‘a population
which was started by a number of sexually derived propagules may thus retain its initial genotypic variation for a very
long period of time, even if it later reproduces almost exclusively asexually’’ (p. 196).
Conclusions
In contrast to the expected decline of genotypic diversity
over time, genotypic diversity, and evenness did not vary between founder and established populations of Cirsium
arvense. On a small geographic scale, sampling at the same
time, we found an extremely high population differentiation
together with a high within-population variability. We interpret these patterns of genotypic and genetic diversity as the
result of the particular reproductive system of the species.
Founder effects and early selection in the seedling stage may
contribute to genetic differentiation among populations.
Then a combination of recruitment of sexually outcrossed
seedlings in the early stages of succession and clonal reproduction allows the species to perform efficient colonization
and to persist locally. The diversity of genotypes as well as
genetic diversity are maintained through time and mainly
seem to reflect the status built up during the early stage of
succession. This strategy appears to be very efficient in C. arvense, and it may have contributed to the worldwide success
of this species.
Literature Cited
Aarssen LW, R Turkington 1985 Within-species diversity in natural
populations of Holcus lanatus, Lolium perenne and Trifolium
repens from four different aged pastures. J Ecol 73:869–886.
Aspinwall N, T Christian 1992 Clonal structure, genotypic diversity,
and seed production in populations of Filipendula rubra (Rosaceae)
from the north central United States. Am J Bot 79:294–299.
Auge H, B Neuffer, F Erlinghagen, R Grupe, R Brandl 2001 Demographic and random amplified polymorphic DNA analyses reveal
high levels of genetic diversity in clonal violet. Mol Ecol 10:
1811–1819.
Ayres DR, FJ Ryan 1997 The clonal and population structure of
a rare endemic plant, Wyethia reticulata (Asteraceae): allozyme and
RAPD analysis. Mol Ecol 6:761–772.
Bakker D 1960 A comparative life-history study of Cirsium arvense
(L.) Scop. and Tussilago farfara L., the most troublesome weeds in
the newly reclaimed polders of the former Zuiderzee. Pages
SOLÉ ET AL.—GENOTYPIC AND GENETIC DIVERSITY IN CIRSIUM ARVENSE
205–222 in JL Harper, ed. The biology of weeds. Blackwell
Scientific, Oxford.
Barrett SCH, JR Kohn 1991 Genetic and evolutionary consequences
of small population size in plants: implications for conservation.
Pages 3–10 in D Falk, K Holsinger, eds. Genetics and conservation
of rare plants. Oxford University Press, Oxford.
Baur B, B Schmid 1996 Spatial and temporal patterns of genetic
diversity within species. Pages 169–201 in KJ Gaston, ed. Biodiversity: a biology of numbers and difference. Blackwell, Oxford.
Bengtsson BO 2003 Genetic variation in organisms with sexual and
asexual reproduction. J Evol Biol 16:189–199.
Bostock SJ, RA Benton 1979 The reproductive strategies of five
perennial Compositae. J Ecol 67:91–107.
——— 1983 Dry weight costs and establishment of seeds and
vegetative propagules. Acta Oecol 4:61–69.
Bussel JD 1999 The distribution of random amplified polymorphic
DNA (RAPD) diversity amongst populations of Isotoma petraea
(Lobeliaceae). Mol Ecol 8:775–789.
Chung MG, BK Epperson 1999 Spatial genetic structure of clonal
and sexual reproduction in populations of Adenophora grandiflora
(Campanulaceae). Evolution 53:1068–1078.
Correns C 1916 Untersuchungen über Geschlechtsbestimmung bei
Distelarten. Sitzungber K Preuss Akad Wiss 20:448–477.
De Riek J, J Dendauw, M Mertens, M De Loose, J Heursel, E Van
Bockstaele 1999 Validation of criteria for the selection of AFLP
markers to assess the genetic variation of breeders’ collection of
evergreen azaleas. Theor Appl Genet 99:1155–1165.
Doyle JJ, JL Doyle 1988 Isolation of plant DNA from fresh tissue.
Focus 12:13–15.
Eber S, R Brandl 2003 Regional patch dynamics of Cirsium arvense
and possible implications for plant animal interactions. J Veg Sci 14:
259–266.
Eckert CG, SCH Barrett 1993 Clonal reproduction and patterns of
genotypic diversity in Decodon verticillatus (Lythraceae). Am J Bot
80:1175–1182.
Ellstrand NC, ML Roose 1987 Patterns of genotypic diversity in
clonal plant species. Am J Bot 74:123–131.
Endler JA 1986 Natural selection in the wild. Princeton University
Press, Princeton, N.J.
Eriksson O 1992 Evolution of seed dispersal and recruitment in
clonal plants. Oikos 63:439–448.
——— 1993 Dynamics of genets in clonal plants. TREE 8:
313–316.
Excoffier L, PE Smouse, JM Quattro 1992 Analysis of molecular
variance inferred from metric distances among DNA haplotypes:
application to human mitochondrial DNA restriction data. Genetics
131:479–491.
Fischer M, R Husi, D Prati, M Peintinger, M van Kleunen, B
Schmid 2000 RAPD variation among and within small and large
populations of the rare clonal plant Ranunculus reptans (Ranunculaceae). Am J Bot 87:1128–1137.
Fischer M, D Matthies 1997 Mating structure and inbreeding and
outbreeding depression in the rare plant Gentianella germanica
(Gentianaceae). Am J Bot 84:1685–1692.
——— 1998 RAPD variation in relation to population size and plant
fitness in the rare Gentianella germanica (Gentianaceae). Am J Bot
85:811–819.
Gabrielsen TM, C Brochmann 1998 Sex after all: high levels of
diversity detected in the arctic clonal plant Saxifraga cernua using
RAPD markers. Mol Ecol 7:1701–1708.
Gibson JP, N Wheelwright 1995 Genetic structure in a population of
a tropical tree Ocotea tenera (Lauraceae): influence of avian seed
dispersal. Oecologia 103:49–54.
Gitzendanner MA, PS Soltis 2000 Patterns of genetic variation in
rare and widespread plant congeners. Am J Bot 87:783–792.
443
Gray AJ 1987 Genetic change during succession in plants. Pages
273–293 in AJ Gray, DE McCauley, PJ Edwards, eds. Colonization,
succession and stability. Blackwell Scientific, Oxford.
Hamrick JL, MJ Godt 1990 Allozyme diversity in plant species.
Pages 43–63 in AHD Brown, MT Clegg, AL Kahler, BS Weir, eds.
Plant population genetics, breeding and genetic resources. Sinauer,
Sunderland, Mass.
Hartl DL, AG Clark 1989 Principles of population genetics. Sinauer,
Sunderland, Mass.
Hartnett DC, FA Bazzaz 1985 The genet and ramet population
dynamics of Solidago canadensis in an abandoned field. J Ecol 73:
407–413.
Hartnett DC, BB Hartnett, FA Bazzaz 1987 Persistence of Ambrosia
trifida populations in old fields and responses to successional
changes. Am J Bot 74:1239–1248.
Heimann B, GW Cussans 1996 The importance of seeds and sexual
reproduction in the population biology of Cirsium arvense:
a literature review. Weed Res 36:493–503.
Ivey CT, JH Richards 2001a Genetic diversity of everglades sawgrass, Cladium jamaicense (Cyperaceae). Int J Plant Sci 162:
817–825.
——— 2001b Genotypic diversity and clonal structure of everglades
sawgrass, Cladium jamaicense (Cyperaceae). Int J Plant Sci 162:
1327–1335.
Jump AS, FI Woodward, T Burke 2003 Cirsium species show
disparity patterns of genetic variation at their range-edge, despite
similar patterns of reproduction and isolation. New Phytol 160:
359–370.
Lalonde RG, BD Roitberg 1994 Mating system, life-history, and
reproduction in Canada thistle (Cirsium arvense; Asteraceae). Am J
Bot 81:21–28.
Maddox DG, RE Cook, PH Winberger, S Gardescu 1989 Clone
structure in four Solidago altissima (Asteraceae) populations:
rhizome connections within genotypes. Am J Bot 76:318–326.
McClintock KA, MJ Waterway 1993 Patterns of allozyme variation
and clonal diversity in Carex lasiocarpa and C. pellita (Cyperaceae).
Am J Bot 80:1251–1263.
McLellan AJ, D Prati, O Kaltz, B Schmid 1997 Structure and analysis
of phenotypic and genetic variation in clonal plants. Pages 185–210
in H De Kroon, J Van Groenendael, eds. The ecology and evolution
of clonal plants. Backhuys, Leiden.
McNeilly T, ML Roose 1984 The distribution of perennial ryegrass
genotypes in swards. New Phytol 98:503–513.
Moore RJ 1975 The biology of Canadian weeds. 13. Cirsium arvense
(L.) Scop. Can J Plant Sci 55:1033–1048.
Pannell JR, B Charlesworth 1999 Neutral genetic diversity in
a metapopulation with recurrent local extinction and recolonization. Evolution 53:664–676.
Piquot Y, P Soumitou-Laprade, D Petit, P Vernet, JT Epplen 1996
Genotypic diversity revealed by allozymes and oligonucleotide
DNA fingerprinting in French populations of the aquatic macrophyte Sparganium erectum. Mol Ecol 5:251–258.
Rousset F 1997 Genetic differentiation and estimation of gene
flow from F-statistics under isolation by distance. Genetics 145:
1219–1228.
Sackville Hamilton NR, B Schmid, JL Harper 1987 Life-history
concepts and the population biology of clonal organisms. Proc R
Soc Lond B 232:35–57.
Schmid B 1990 Some ecological and evolutionary consequences
of modular organization and clonal growth in plants. TREE 4:
25–34.
Schmidt K, K Jensen 2000 Genetic structure and AFLP variation of
remnant populations in the rare plant Pedicularis palustris
(Scrophulariaceae) and its relation to population size and reproductive components. Am J Bot 87:678–689.
444
INTERNATIONAL JOURNAL OF PLANT SCIENCES
Schneider S, D Roessli, L Excoffier 2000 Arlequin version 2.000:
a software for population genetic data analysis. Genetics and
Biometry Laboratory, University of Geneva.
Sherman-Broyles SL, JP Gibson, JL Hamrick, MA Bucher, MJ
Gibson 1992 Comparisons of allozyme diversity among rare and
widespread Rhus species. Syst Bot 17:551–559.
Slatkin M 1977 Gene flow and genetic drift in a species subject to
frequent local extinctions. Theor Pop Biol 12:253–262.
Smith B, JB Wilson 1996 A consumer’s guide to evenness indices.
Oikos 76:70–82.
Taylor DR, LW Aarssen 1988 An interpretation of phenotypic
plasticity in Agropyron repens (Gramineae). Am J Bot 75:
401–413.
Travis SE, J Maschinski, P Keim 1996 An analysis of genetic
variation in Astragalus cremnophylax var. cremnophylax, a critically endangered plant, using AFLP markers. Mol Ecol 5:
735–745.
Verburg R, J Maas, HJ During 2000 Clonal diversity in differentlyaged populations of the pseudo-annual clonal plant Circaea
lutetiana L. Plant Biol 2:646–652.
Vos P, R Hogers, M Bleeker, M Reijans, T van de Lee, M Hornes, A
Frijters, et al 1995 AFLP: a new technique for DNA fingerprinting.
Nucleic Acids Res 23:4407–4414.
Walther-Hellwig K, R Frankl 2000 Foraging habitats and foraging
distances of bumblebees, Bombus spp. (Hym., Apidae), in an
agricultural landscape. J Appl Entomol 124:299–306.
Whitlock MC 1992 Temporal fluctuations in the demographic
parameters and the genetic variance among populations. Evolution
46:608–615.
Whitlock MC, DE McCauley 1990 Some population genetic consequences of colony formation and extinction: genetic correlations
within founding groups. Evolution 44:1717–1724.
Widen B, N Cronberg, M Widen 1994 Genotypic diversity, molecular markers and spatial distribution of genets in clonal plants,
a literature survey. Folia Geobot Phytotaxon 29:245–263.
Williams CB 1964 Patterns in the balance of nature. Academic Press,
London.
Wolf AT, RW Howe, JL Hamrick 2000 Genetic diversity and
population structure of the serpentine endemic Calystegia collina
(Convolvulaceae) in northern California. Am J Bot 87:1138–1146.
Wright S 1965 The interpretation of population structure by
F-statistics with special regards to systems of mating. Evolution
19:395–420.
Xie Z, Y Lu, S Ge, D Hong, F Li 2001 Clonality in wild rice (Oryza
rufipogon, Poaceae) and its implications for conservation management. Am J Bot 88:1058–1064.